PSI - Issue 68

Robert Sundström et al. / Procedia Structural Integrity 68 (2025) 1081–1090 Robert Sundström / Structural Integrity Procedia 00 (2025) 000–000

1085

5

Grinding can give tensile residual stresses at the surface while polishing can give compressive residual stresses and annealing removes these stresses. Tensile residual stresses were observed to be more detrimental for the fatigue life ratio than surface roughness. An interesting observation that lacked explanation was that high cycle fatigue lives of the polished condition were increased in hydrogen gas compared to air. In summary, these studies show that there can be effects from the surface condition on mechanical testing results in hydrogen environments when using a solid specimen tested in an autoclave or under cathodic charging. They can depend on alloy, surface roughness and residual stress state in the surface layer. Microstructural changes, for example formation of martensite or a recast layer caused by electric discharge machining (EDM), must also be considered. 2.3. Surface finish effects for hollow specimens Manufacturing small, long holes with high length-diameter ratios is difficult but can be achieved through deep hole drilling, where cutting fluid is supplied through a hollow in the drill bit. Electrical discharge machining (EDM) can also be used to either manufacture long holes from the start using sink EDM, or to refine the surface of an existing hole by wire EDM. Small holes are necessary because larger hole diameters give lower wall thicknesses that can affect testing results (Ogata and Ono 2019). On the other hand, larger hole diameters are easier to manufacture and simplify post-processing of the inner surface. Machining can leave a surface layer with altered properties compared to the bulk. Wire electric discharge machining (EDM) leaves a recast surface layer (Ogata and Ono 2019). In one study, drilling of a 6 mm diameter hole in 316 LN austenitic stainless steel gave a hardened surface layer of about 200 µm with an average roughness of 2.4 µm (Xiong et al. 2020) . This could have effects on mechanical testing results. Various methods can be used for post processing of holes: reaming, honing, abrasive flow machining, and manual grinding with abrasive paper, coated wires or brushes. The ISO standard for the hollow specimen allows a range of 3 to 12 mm and 1 to 4 mm for the outer and inner diameter respectively, and for a ratio between the inner and outer diameter of less than or equal to 0.33 (ISO 2024). It recommends the removal of any machining layers and a surface condition of average roughness Ra = 0.25 µm but prescribes no method for achieving this. A standard for mechanical testing of solid specimens in gaseous hydrogen environments, ASTM G142-98, prescribes an average surface roughness of 0.25 µm and a removal of 0.05 mm of the diameter in the last two machining passes by grinding (ASTM 2022). In a hollow specimen, this requirement would correspond to a removal of 25 µm of the inner surface. Since surface finish might affect mechanical testing results in hydrogen environments and the hollow specimen has an inner surface that is difficult to polish, the effects of inner surface finish on testing results have been explored by researchers (Ogata and Ono 2019, Michler et al. 2022, Michler et al. 2023, Campari et al. 2024, Konert et al. 2024, Michler et al. 2024). Using a hollow specimen geometry, a low-alloy steel and stainless steel with three different surface conditions was slow strain rate tested at 193, 228 and 300 K: wire electric discharge machined (wire-cut, Rz = 12 µm), grinding by electroplated diamond wire (honing, Rz = 2 µm) and abrasive flow machining (polishing, Rz = 3 µm) (Ogata and Ono 2019). Reduction of area (RA) is a common measure of ductility, and by dividing the value measured after slow strain rate testing in a hydrogen environment, by a value measured after testing in a reference environment such as air or inert gas, the relative reduction of area (RRA) can be obtained. This is a commonly used hydrogen embrittlement index (HEI) for conventional solid smooth specimens (Lee 2016). The study (Ogata and Ono 2019) found the relative reductions of area for the stainless steel to be broadly similar comparing the different surface conditions. The low alloy-steel was more sensitive to surface conditions, with the wire-cut condition having approximately 20% lower relative reductions of area than polished or honed samples at room temperature (300 K). In general, the finer surfaces of the honed and polished samples resulted in higher relative reduction of area, i.e., less embrittlement. There were no effects of surface finish on relative tensile strength at room temperature for either alloy. Overall, the results pointed to honing with a diamond wire as an acceptable post-processing method for evaluation of relative reduction of area and relative tensile strength at room temperature for a stainless steel and a low-alloy steel. The effects at lower temperatures were sometimes different than those at room temperature, so testing temperature also needs to be considered when choosing manufacturing and post-processing methods.